NEW ORLEANS – The use of proton pump inhibitors (PPIs) was associated with significantly increased risk of having a first ischemic stroke in a large nationwide Danish cohort study, Thomas S. Sehested, MD, reported at the American Heart Association scientific sessions.
The relationship was dose dependent. At the lowest available dose of each of the four PPIs studied there was no significantly increased risk. At the intermediate doses of three of the four PPIs studied, the increased risk of ischemic stroke became statistically significant. And the highest dose of each drug was associated with the greatest ischemic stroke risk.
Bruce Jancin/Frontline Medical News
Dr. Thomas S. Sehested
In Denmark, for instance, where most PPIs are prescription only and use is easily trackable, it’s estimated that, at any given time, 7% of the adult population is taking a PPI, often not as directed in the labeling.
The impetus for this study, Dr. Sehested explained, was the mounting evidence that PPIs may constitute an independent risk factor for acute MI and other cardiovascular events. For example, a recent meta-analysis of 17 randomized controlled trials totaling 7,540 participants published through mid-2015 concluded that the use of PPIs was associated with a 70% increase in cardiovascular risk (Neurogastroenterol Motil. 2016 Aug 30. doi: 10.1111/nmo.12926).
He reported on 245,676 Danes above age 30 who were free of prior MI or stroke when they underwent elective GI endoscopy during 1997-2012. After a 30-day postendoscopy grace period during which 1,476 patients had a first MI, stroke, or died of any cause, the final study population was 244,200, of whom 43.7% were PPI users during the grace period and beyond.
During a median 5.8 years of follow-up, 9,489 subjects (3.9%) had a first ischemic stroke. Because of the comprehensive nature of Denmark’s interlocking birth to death registries, there was virtually no loss to follow-up in this study.
The unadjusted incidence of ischemic stroke in PPI nonusers was 55.7 per 10,000 person-years, compared with 88.9 per 10,000 in PPI users.
The PPI users were slightly older than nonusers by roughly 3 years. They were also an absolute 5% more likely to be hypertensive and an absolute 1.7% more likely to be regular users of NSAIDs. All of these differences, while modest, were statistically significant because of the large patient numbers involved.
In a multivariate analysis adjusted for age, sex, calendar year, comorbid diabetes, hypertension, alcohol use disorder, heart failure, peptic ulcer, peripheral artery disease, kidney disease, aspirin, oral anticoagulants and other medications, and socioeconomic status, current users of PPIs were 19% more likely to have a first ischemic stroke than nonusers. That difference is statistically significant and clinically meaningful, Dr. Sehested said.
In contrast, when the same sort of nationwide analysis was repeated, comparing current users of histamine-2 receptor antagonists to nonusers of those drugs or PPIs, there was no difference in ischemic stroke risk between the two groups.
The message, according to Dr. Sehested, is that physicians should encourage more cautious use of PPIs. And especially in the United States, where most PPIs are available over the counter, it’s prudent during office visits to ask what nonprescription drugs a patient is taking.
Dr. Sehested presented his study findings in a session devoted to original research in cardiovascular epidemiology. Many top American epidemiologists were present in the audience, and several rose to congratulate him on his presentation of the latest elegant epidemiologic study to come out of Denmark, the only place in the world where this sort of nationwide comprehensive research is possible.
“Wow! I just love the work you do in Denmark. It’s really inspiring,” commented David Siscovick, MD, senior vice president for research at the New York Academy of Medicine and professor emeritus of medicine and epidemiology at the University of Washington in Seattle.
He had a question: “Did you deal with PPI starters and stoppers and compliance in any way?”
Dr. Sehested replied that he and his coinvestigators were able to see who was on a PPI at any given point in the study, and they accounted for that. One issue the researchers plan to examine but haven’t yet had a chance to, however, is the relationship between duration of PPI therapy and ischemic stroke risk. It’s likely that some patients had already been on a PPI for a lengthy time at elective endoscopy, which is when the study in its current form began.
“I think that would strengthen the study,” he said.
Comoderator Jorge Kizer, MD, of Albert Einstein College of Medicine in New York, commented, “Confounding by indication is clearly the elephant in the room. The guidelines actually recommend adding a PPI if a patient is on dual-antiplatelet therapy and has an NSAID added. Did you adjust for that? It would boost confidence that the results are actually due to the PPI.”
Dr. Sehested answered that the great majority of individuals with cardiovascular disease at baseline were excluded from the analysis.
“I don’t think we had that many on dual-antiplatelet therapy,” he added.
Preclinical studies suggest a possible mechanism by which PPIs may harm cardiovascular health. The drugs reduce nitric oxide synthase levels, with resultant endothelial dysfunction, he said.
Dr. Sehested is employed at the Danish Heart Foundation, which funded the study.